A communication technique allowing persons respectively in a plurality of distant spaces to have conversation while viewing their video images each other has been already known. Such conversation communication is implemented in such a manner that communication units provided respectively in the plurality of distant spaces communicate with each other.
In order to terminate the above-described conversation communication, one of the communicators takes action (hereinafter referred to as “termination action”) such that a predetermined termination condition is satisfied. In a typical conversation communication system, output of words or a gesture as a sign for termination of conversation is employed as the termination action (see, e.g., JP 2014-090230 and JP 2010-183205). As an example, in communication systems described in Patent Literatures 1 and 2, output of words such as “good bye” or a gesture such as handwaving from one of communicators is detected as a sign for termination of conversation. Then, conversation communication is interrupted with detection of action, i.e., the sign for termination of conversation, as a trigger.
In the case of using words or a gesture as the sign for termination of conversation as in JP 2014-090230 and JP 2010-183205 described above, many words and gestures as described above need to be assumed and be registered in advance in order to accept various communicators and conversation situations (e.g., a spoken language and a gesture vary according to a communicator's attribute and a relationship between communicators). However, in fact, it is extremely difficult to assume all situations and register words and gestures in advance. On the other hand, when the communicator involuntarily outputs the words or the gesture registered as the sign for termination of conversation, the conversation communication is interrupted against the intention of wishing to continue conversation.
Alternatively, it may be configured such that a human sensor is provided in a space (a room) where the communicator is present while having conversation and that a condition where the human sensor no longer detects a person in the space is taken as the conversation termination condition. Note that in this configuration, a person still remaining in the space after the communicator has moved to the outside of the space is erroneously detected by the human sensor. In such a case, the conversation communication is continued although conversation has ended.
As another alternative, it may be configured such that a voice detection sensor is provided in the space (the room) where the communicator is present while having conversation and that a condition where the voice detection sensor no longer detects voice in the space is taken as the conversation termination condition. Note that in this configuration, noise caused in the space after the communicator has moved to the outside of the space is, as in the above-described configuration, erroneously detected by the voice detection sensor. As a result, the conversation communication is continued although conversation has ended.